[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bpfk] official cmavo form

John Cowan, On 21/10/2014 19:11:

And Rosta scripsit:


It sounds like there are all these exceptions because the rule is
wrongly formulated. If syllables are discarded and the metrical units
are instead AEIOU clusters, might the rule become exceptionless?


It would get worse:  we would have to have an exception for words like /stabaa/
explaining why it is /sta'baa/ rather than /'stabaa/.


Is /stabaa/ an alternative notation or analysis for /staba'a/? If not, what is it? If it is, then we just need a suitable definition of AEIOU cluster: either we say that /'/ is phonologically visible, in which case it can't be cluster-internal (since it is not an AEIOU), or we take 'AEIOU cluster' as a primitive rather than derived term, and have a rule of apostrophe-insertion between phonologically distinct but contiguous AEIOU clusters.


Obviously it was the glideless /ae, ea, aa/ type that led to Lojban's
"'". That in itself was not so bad a move, tho the choice of realization
was, but making it contrastive with zero between other vowels gives
greater headaches. I'd have just forbidden them altogether; going all
Livagian on their ass, I'd allow i to be followed by any vowel but i,
u to be followed by any vowel but u, e to be followed by no vowel but i,
o to be followed by no vowel but u, and a to be followed by no vowel
but i and u.


This would, of course, involve a complete discarding of the cmavo list and
starting over.


A certain revision, rather than a complete discarding. You could convert to new forms by rule, /e'V/ to /eiV/, /o'V/ to /ouV/, /i'V/ to /iV/, /u'V/ to /uV/, some other rule for /a'V/, and sort out the newly created homophones, perhaps by making use of a /aiV/:/auV/ contrast.

But to be more realistic and hence more conservative, my reading of what xorxes said camxes does, namely make every string analysable as a sequence of CVs, sounds like the best rule.

John Cowan, On 21/10/2014 19:08:> And Rosta scripsit:



Is the concern that because /./ is elidable when its presence is not
morphologically contrastive, the risk is that through habit it would
end up being elided even when it is?


Just so.


a problem with that is that it is hard to carefully and deliberately
show that one is using a properly /./-less form.


Indeed.


This is a wider problem with /./, isn't it. A solution would be to make glottal-stop elision illicit.

In any case, I was talking about "a ua" [a?wa] as hard to distinguish

>from "a'ua" [ahwa], both tending to become simple [awa].

Specifically for L1 English speakers, you must mean, rather than for
people in general. Does it really make sense to base the rules of
Lojban on the specific needs of L1 English speakers?


Both [W] and the cluster [hw] are rare in the world's languages compared
with [w], so it's not too surprising that most varieties of English have
lost them.


I think [hw] is virtually inarticulable. I don't know if anybody knows the frequency of [W] or any other phone in the world's languages. The fact that it's rare as the primary allophone of a phoneme doesn't mean it's rare as a phone. (E.g. bilabial trills are rare as primary allophone of a phoneme in world's languages, but the phone is still to be heard in English words for quite a few speakers.)

But anyway, rather than [aWa] tending to become [awa] and hence neutralized with /aua/, it could instead become the far more innocuous [aWua].


Many L1 English speakers would tend to hear /a.ua/ as /at ua/.


Lojban /t/ is problematic for anglophones in general, given the North
American (i.e majority) tendency to voice it between vowels and to
glottalize it between a vowel and a syllabic consonant.  What is worse,
all anglophones tend to hear [t] (as opposed to [t_h]) as /d/.


(Most but not all.)


I don't think we can do anything about this.


We could specify aspirated realizations for /ptk/ and voiced for /bdg/.

If /'/ is to be kept distinct from /x/, /'/ must be [T], giving [aTua]
for /a'ua/, which is unlikely to become [awa].


I'm not sure if this is meant to be an anglophone or a universal claim.


A universal phonetic claim.


Anglophones tend to render [x] as [k], as in _loch, bach, Bach_, and
Germans have no problem distinguishing /h/ and /x/ systematically,
though it's arguable that there are no [h] : [x] minimal pairs, as [h]
is only in onsets whereas /x/ in onsets is realized (in the standard
accent, at least) as /C/.


For phonetic reasons, [h] is possible only between vowels sufficiently open that the narrowest constriction of the vocal tract is at the glottis, so e & a but not i, u, o. Frication (turbulence) occurs at the locus of narrowest constriction.

Any posterior fricative will tend strongly to assimilate to [C] in environment [i _ i] and to [W] (or labialized [x_w]) in environment [u _ u].


It would be an assimilation rather than a fortition. As I've said
before, [h] is articulatorily impossible as a realization of /'/
in some environments, e.g. /i'i/, at ordinary speech rates,


I articulate /i'i/ as [iCi], /u'u/ as [uWu], /ii/ as /j\i/ (with a voiced
palatal fricative like Spanish-Spanish "y"), and /uu/ as [wu].


How about /ixi/ and /uxu/?

--And.

--
You received this message because you are subscribed to the Google Groups "BPFK" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bpfk-list+unsubscribe@googlegroups.com.
To post to this group, send email to bpfk-list@googlegroups.com.
Visit this group at http://groups.google.com/group/bpfk-list.
For more options, visit https://groups.google.com/d/optout.